Abstract:
본 발명은 레이드와 같이 스트라이핑 기법으로 다수개의 디스크로 구성된 디스크어레이에서의 순차적 프리페칭 방법에 관한 것으로, 본 발명은 프로세스로부터 읽기 명령을 수신하는 제1 과정과, 상기 수신된 읽기 명령이 지시하는 초기 프리페칭 크기와 프리페칭할 연속된 블록들을 결정하는 제2 과정과, 상기 연속된 블록들이 서로 다른 스트립에 포함되는지를 확인하는 제3 과정과, 상기 연속된 블록들이 서로 다른 스트립에 포함되면, 상기 연속된 블록들 중 시작 블록이 속해있는 스트립에 포함된 블록들의 수만큼 상기 초기 프리페칭 크기를 줄이는 제4 과정과, 상기 줄어든 초기 프리페칭 크기에 따라 하나의 디스크 명령을 생성하는 제5 과정을 포함을 포함함을 특징으로 한다. 상술한 본 발명은 스트라이핑 기법을 사용하는 디스크 저장장치에서 독립성 손실로 인한 성능 저하를 방지할 수 있는 효과가 있다. 저장장치 관리, 프리페칭, 디스크어레이
Abstract:
An apparatus and a method for detecting a location by using a synchronously distinguishable marker are provided to remove error due to signal disturbance to reduce error rate and divide markers into synchronous signals according to time, thereby preventing an increase of installation costs due to image recognition and a decrease of error rate. A light emitting module(10) is formed in the same shape. The light emitting module is arranged at the same interval and is on/off at a predetermined interval. An emitting module driving unit(20) controls on/off control of each light emitting marker comprising a light emitting module. A synchronous signal for time synchronization is wirelessly transmitted to a location recognition movement terminal(30) so that the location recognition movement terminal recognizes a location of a light emitting marker which is on at a specific time.
Abstract:
An intuitive real spatial aiming-based system and identification and communication methods are presented. The intuitive real spatial aiming-based system includes aiming device, target devices and an indoor location-based service server. The aiming device includes a mobile computer or a Personal Digital Assistant (PDA), which is provided with a first wireless communication tool, a first location tracking sensor, and an electronic compass having a magneto resistive sensor for detecting a directions. The target devices each includes a fixed computer, a home appliance, a PDA, or a mobile communication terminal, which is selectively provided with second wireless communication tool and a second location tracking sensor. The indoor location-based service server tracks the indoor locations and coordinates the aiming device and the target devices in real time. To tolerate sensor's error, an angle-based target region, a width-based target region, and a combination of the two regions presented.
Abstract:
선택적으로 패터닝되는 타일 마스크 셋 및 마스킹 마스크를 이용하는 스트럭처드 ASIC의 레이어 리소그래피 방법이 개시된다. 스트럭처드 ASIC의 레이어를 리소그래피하기 위하여, N개의 마스크 쌍들을 결정하며, 상기 N개의 마스크 쌍 중 타겟 마스크 쌍을 선택한다. 상기 타겟 마스크 쌍의 마스킹 마스크를 통해 1차 노광하고, 상기 타겟 마스크 쌍의 타일 마스크 셋을 통해 2차 노광한다. 상기 1차 노광 및 2차 노광된 웨이퍼를 식각하여 레이어를 선택적으로 리소그래피한다. 따라서, 디자인의 특성에 맞게 복수의 타일들을 적절히 배치하여 종래의 스트럭처드 ASIC의 비용 절감 효과를 유지하면서 성능을 향상시킬 수 있다.
Abstract:
PURPOSE: A layer lithographic method of a structured application specific integrated circuit(ASIC), a design method, and a tile mask set used in the same are provided to improve performance while maintaining cost reduction effects of an ASIC using a selectively patterned tile mask set and a masking mask. CONSTITUTION: First to N-th mask pairs are determined(S110). A target mask pair is selected among the first to N-th mask pairs(S120). A second tile part is first exposed using a masking mask of the target mask pair(S130). A first tile part is second exposed using a tile mask set of the target mask pair(S140). The second exposed first tile part is etched(S150). A lithography state of a pattern with respect to all titles is determined(S160).
Abstract:
본 발명은 프리페칭 데이터의 적응적 메모리 관리 방법에 관한 것이다. 종래의 프리페칭 기법은 프리페칭의 정확성을 중요시한다. 왜냐하면 잘못된 예측으로 프리페칭된 사용되지 않는 데이터는 캐시를 오염시키기 때문이다. 그러나 만약 새로운 방식이 프리페칭 적중률과 캐시 적중률의 합이 어떤 프리페칭의 프리페칭 적중률과 프리페칭 없는 캐시 적중률 중 어느 것보다도 높으면, 낮은 프리페칭 정확성의 문제는 극적으로 해결될 수 있다. 이 목적을 이루고자, 적응적 캐시 솎아내기를 제안한다. 이것은, 프리페칭 적중률과 캐시 적중률의 두 변경(邊境) 사용률들을 동일하도록 프리페칭 블록 캐시들의 수와 캐싱된 블록 캐시들의 수를 최적으로 할당함으로써, 적응적 방법으로 적당한 시간에 사용되지 않는 프리페칭된 블록들을 메모리에서 퇴출시킨다. 프리페칭, 디스크어레이, 퇴출, 솎음
Abstract:
PURPOSE: A layer lithography method for a structured ASIC(Application Specific Integrated Circuit), a designing method, a selectively patterned mask set, and a masking mask are provided to reduce manufacturing costs by selecting and arranging a plurality of tiles according to the design property by using the selectively patterned mask and masking mask. CONSTITUTION: M mask sets are made for lithographing a layer of a structured ASIC(S10). N masking masks are made for lithographing the part of mask sets on a wafer according to the previously designed structured ASIC(S20). The part of the first mask set is selectively lithographed on a wafer by using the first mask pair(S40). The part of the second to N-th mask sets is selectively lithographed on the wafer by using the second to N-th mask pairs(S50).
Abstract:
PURPOSE: A sequential pre-fetching method in a computer system is provided to prevent performance from being lowered in consideration of independence during pre-fetching in a computer storage device using disks. CONSTITUTION: A sequential pre-fetching method in a computer system comprises the following steps. It is checked whether consecutive blocks are included in at least two strips. When the consecutive blocks are included in at least two strips, a pre-fetching size is reduced to include blocks belonging to one pre-fetching request only in one strip. A disk command is generated according to the beginning of consecutive blocks to be pre-fetched and the reduced pre-fetching size.
Abstract:
A pre-fetching data managing method in a computer storage device is provided to increase the efficiency in consideration of the pre-fetching. The entire cache is managed by a strip cache unit. Strip caches are divided to an upstream and a downstream. The upstream has a pre-fetched block cache and a cached block cache(305). The downstream has only the cached block cache(306). According to the processing result, the number(Nu) of strip caches which the upstream can have is updated by using the differential value of the sum of a pre-fetching hit rate and a cache hit rate(309). When the number of strip caches included in the upstream is bigger than the updated number(310), a strip cache of the upstream which is not used for the longest time is moved to the downstream according to the LRU(Least Recently Used) policy(311).
Abstract:
A method for improving input/output performance of an RAID(Redundant Array of Independent Disks) system using an MSC(Matrix Stripe Cache) is provided to instruct disks to perform read and write operations based on an rxw matrix to which additional read and write operations are added to reduce the number of read and write inputs and output, thereby improving write performance. A method for improving input/output performance of an RAID system using an MSC includes a first step of generating an rxw matrix corresponding to a read/write matrix before row combination from a unit MSC which performs a write operation on a disk, a second step of generating a row-combined rxw matrix corresponding to a final read/write matrix through row combination from the rxw matrix generated in the first step, and a third step of performing a read operation based on the row-combined rxw matrix, carrying out an XOR operation on each row of the row-combined rxw matrix, and performing a write operation on each column of the row-combined rxw matrix.